"... In PAGE 5: ...( Table3 ). These effects were modest compared to the deffs that incorporate weighting as well as clustering, indicating that most of the increase in variance was due to the sample weights.... ..."

"... In PAGE 4: ...transpiration of tules in the Central Valley is about 2642 mm of water annually and the transpiration of cattails is about 2286 mm annually (Young and Blaney 1942). Riparian forest, composed of medium tall to tall broad-leaved deciduous trees, occupied natural river levees and accounted for approximately 8% of the total area of the Central Valley ( Table1 ). The composition of the native riparian forests is uncer- tain (Griffin 1977), but Thompson (1961) provides an excellent review of historical accounts and Kuch- ler (1977) suggests that cottonwood (Populus frernontii) was the dominant species.... In PAGE 4: ... The native vegetation in the Central Valley in- cluded two mesophytic vegetation types, the Cali- fornia prairie and the valley oak savanna (Kuchler 1977). The pristine California prairie accounted for the largest area of pre-agricultural vegetation in the region ( Table1 ). It covered about 63% of the Cen- tral Valley and was over four times larger in area than the tule marsh which was the next largest vege- tation type.... In PAGE 5: ... Although consumptive use by native vegetation on drying soil varies within a given vegetation type in response to local climate, soil, and drainage differences (Miller 1977), data for a representative site provide a reasonable basis for estimating the areal moisture flux. Representative evapotranspiration from meso- phytic vegetation is lower than evapotranspiration for hydrophytes and riparian vegetation by as much as a factor of eight ( Table1 ). The significantly low- er evapotranspiration flux for these plants is the product of differences in plant physiology (Linacre 1976) and the reduced availability of moisture to plants during the dry summer months.... In PAGE 5: ... Annual eva- potranspiration from the valley grassland is esti- mated to be 282 mm since it occurred in areas where precipitation was moderately limiting (Major 1977). The other mesophytic vegetation type, valley oak savanna, covered 204,000 ha and accounted for the smallest area of native vegetation in the Central Valley ( Table1 ). This vegetation type is character- ized by valley oak (Quercus lobata) growing in groups or singly and towering above an underlying California prairie cover.... In PAGE 5: ... The evapotranspiration for this mesophytic vegetation is estimated to be 600 mm which com- bines the limited moisture available to the grasses and the unlimited moisture available to the scat- tered oaks (MacGillivray 1975). San Joaquin saltbrush is the fifth major native vegetation type in the Central Valley ( Table1 ). This xerophytic vegetation commonly occupied poorly drained alkali soils in the arid southern and western portions of the Tulare subbasin and a comparative- ly small area of the central San Joaquin subbasin.... In PAGE 5: ... The large ex- panse of mesophytic California prairie accounted for another 25% of the total evapotranspiration flux. Although the tule marsh was the dominant source of evapotranspiration for the Delta and Tulare subbasins, its contribution to total evapo- transpiration was slightly less in the San Joaquin subbasin and lower still in the Sacramento subbasin ( Table1 ). The larger areas of riparian forest and California prairie in the Sacramento and San Joa- quin subbasins compensated for the higher rate of evapotranspiration by the tule marsh.... In PAGE 6: ... This agricultural land was previously forests, marsh and overflow land, grasslands, and alkali flats (Harding 1960; Thompson 1961). The greatest expanse of surviving native vegeta- tion is in the Sacramento subbasin and the smallest area is in the San Joaquin subbasin ( Table1 ). The greatest conversion of land from native vegetation to irrigated agriculture has occurred in the Tulare subbasin where native vegetation has been reduced to about 27% of its pre-agricultural area.... In PAGE 6: ... The greatest conversion of land from native vegetation to irrigated agriculture has occurred in the Tulare subbasin where native vegetation has been reduced to about 27% of its pre-agricultural area. California prairie continues to be the most ex- pansive native vegetation throughout the Central Valley even though its area has been reduced to about 46% of its original coverage ( Table1 ). The present-day prairie occurs as a ring around the Cen- tral Valley that increases in size from south to north.... In PAGE 6: ... San Joaquin saltbrush has been replaced by ir- rigated crops on all but the most alkaline soils in the arid southern end of the Central Valley. Today, saltbrush occupies about one-fourth the area of its pre-agricultural habitat, but nearly 90% of the salt- brush is in the Tulare subbasin ( Table1 ). The rela- tively flat valley floor in the San Joaquin and Tulare subbasins occupied by San Joaquin salt- brush was easily converted to agriculture as soon as irrigation water became available.... In PAGE 6: ... The irrigated acreage in these subbasins today accounts for more than one-half of the irrigated land in California (McCorkle and Nuckton 1983). Irrigated agriculture has nearly eliminated the valley oak savanna from is native areas in the Tulare subbasin ( Table1 ). Although valley oak savanna accounted for the smallest area of the five native vegetation types in the Central Valley, it has experienced the greatest relative reduction in area.... In PAGE 9: ... In general, na- tive vegetation has been replaced less extensively in the northern section of the valley than in the southern portion. The present landscape in the Sacramento subbasin has the largest area of native vegetation and the Delta subbasin has the third lar- gest area ( Table1 ). Although the Tulare subbasin has the largest total land area, its present area of na- tive vegetation is only 60,000 ha larger than the areal expanse of native vegetation in the Delta sub- basin.... In PAGE 9: ... This does not suggest, however, that irrigated agriculture has not altered the landscape in this sub- basin. A significant component of the change in na- tive vegetation has been a 52% reduction in tule marsh ( Table1 ) which has resulted from the crea- tion of islands protected by levees. The cultivated land on the leveed islands was previously covered by water for most or all of the year.... In PAGE 9: ... The Tulare subbasin represents the most severe change in vegetation resulting from the introduc- tion of irrigated agriculture. Only about 27% of the subbasin area is still inhabited by native vegetation ( Table1 ). Irrigated crops occupy all of the land that was once riparian forest and all but a small percen- tage of the valley oak savanna habitat.... In PAGE 9: ... Changing the vegetation structure of the Central Valley has altered the transfer of moisture to the at- mosphere, but there are several surprising facets to the change in evapotranspiration. In the pre- agricultural landscape, the largest evapotranspira- tion flux was provided by the Tulare subbasin and was supported largely by the expanse of tule marsh ( Table1 ). The arid conditions in this subbasin make such a large volume of evapotranspiration unex- pected.... In PAGE 9: ... The change in evapotranspiration in the Delta subbasin is notable also. In this case, evapotranspi- ration for the irrigated landscape is less than evapotranspiration for the pre-agricultural land- scape by about 1,000 hm3 ( Table1 ). This is the result of tule marsh and riparian forest being replaced by irrigated crops which use less water than the native vegetation.... ..."

"... In PAGE 14: ... Only grain was consumed by all households. Table2 provides the MSLE of the LES with demographic e ects. In solving the MSLE, the rst concern is that the estimated i may not satisfy the restriction xi? i gt; 0: To satisfy this constraint, the reparameterizations of i = min(xi) ? exp( i ); where min is taken over the whole sample and i are free parameters, are used: A second concern is that the estimated var-cov matrix may not be positive de nite; therefore, positive de niteness is imposed by replacing with the product of a lower triangular matrix, i.... In PAGE 14: ... lower triangular matrix, i.e., = AA0; A = [aij] ; aij = 0; i lt; j: Third, because the LES price and expenditure elasticities are functions of all prices and expenditures rather than constants, these elasticities do not provide a transparent summary of the behavior implied by a particular set of LES parameters. Since the demand for the ith good in the LES is inelastic if i is positive and is elastic if i is negative, we can determine from Table2 that demand for all seven goods is elastic. Finally, we use the likelihood ratio statistics to test the signi cance of the demographic variables.... In PAGE 14: ... Demographic variables do matter. From Table2 it is clear that family size signi cantly a ects consumption patterns. Lee and Pitt (1986b) imposed a restrictive assumption that error terms in the LES are independent.... ..."

"... In PAGE 6: ... (5). We obtained these coefficients ( Table1 ) from the data of Huang and Warner (1995) and Huang and Nanson (1998), who derived bank strength indices for various bank types with different levels of vegetation cover and related them to sc. Millar and Quick (1998) reported similar coefficients for sc between different bank vegetation types.... In PAGE 6: ... sinense Lour.), we used sc coeff for bsparseQ and bdenseQ trees ( Table1 ). For channel banks with English ivy (H.... ..."

"... In PAGE 5: ... The resulting vegetation map was based on the interpretation of 1:12,500 scale airphotos (collected 1982) and a subsequent intensive field survey (carried out in 1993) (see Michalsky amp; Ellis, 1994). This map subdivided GNP into seven discrete vegetation land units (VLUs), which were further classified according to vegetation type (VT) ( Table3 ). We then assigned C4 cover values to each VT.... In PAGE 12: ... This confusion is likely a result of two factors. First, tree communities make up less than 1% of the total area of GNP ( Table3 ). As a result, the discriminant function defined for this class is based on few observations.... ..."

"... In PAGE 4: ...eaf Water (0.005 - 0.02 g/m2), and LAI.(0.5 - 6 m2/m2)). The sun zenith angle is varied between 0 and 48 degrees. Figure 2 displays the change of the parameters in terms of absolute reflectance as described in Table2 . Finally Figure 3 lists the absolute delta reflectances for all vegetation variables and radiometric situations.... ..."

"... In PAGE 14: ...as been used with a learning rate of 0.1 and momentum of 0.9. The number of samples used for analysis is shown in Table1 for vegetation data and in Table 2 for natural object data. The co-occurrence matrix method fails to work with very small regions generated by the segmentation process.... ..."

"... In PAGE 7: ... This frustrates any attempts at verifying, comparing, or aggregating the results of different studies. The problem is illustrated in Table21 which shows the effect, the measurement technique, and the plant investigated. Table 21.... ..."